Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Information extraction from Korean radiology reports mingled two language.

Miyoung Kwak1, Seungbin Han, Jinwook Choi

  • 1Department of Biomedical Engineering, College of Medicine, Seoul National University.

AMIA ... Annual Symposium Proceedings. AMIA Symposium
|June 17, 2006
PubMed
Summary

This study introduces Concept Nodes (CNs) for effective information extraction from Seoul National University Hospital (SNUH) radiology reports. The method successfully processes mixed Korean and English text, demonstrating its utility in clinical data analysis.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Clinically grounded retinal representation learning from minimal supervision.

Scientific reports·2026
Same author

Proteome profiling reveals NQO2 activity contributing to proteasome inhibitor resistance in multiple myeloma cell lines.

Molecular & cellular proteomics : MCP·2026
Same author

Early fibrotic niches establish tumour-permissive microenvironments.

Nature·2026
Same author

Promoter hypomethylation drives ABCB1-mediated carfilzomib resistance in multiple myeloma.

Clinical epigenetics·2026
Same author

Loss of GPRC5D enhances the proliferative capacity and competitive fitness of myeloma upon anti-GPRC5D immunotherapy.

Leukemia·2026
Same author

scMILD: Single-cell multiple instance learning for sample classification and associated subpopulation discovery.

iScience·2026

Area of Science:

  • Medical Informatics
  • Natural Language Processing
  • Clinical Data Mining

Background:

  • Radiology reports contain crucial clinical information.
  • Processing multilingual clinical text presents significant challenges.
  • Existing information extraction methods may struggle with mixed-language data.

Purpose of the Study:

  • To develop and evaluate an Information Extraction (IE) system for Seoul National University Hospital (SNUH) radiology reports.
  • To assess the effectiveness of Concept Nodes (CNs), a case frame-based extraction rule, for processing coexisting Korean and English text.
  • To investigate the utility of syntactic and semantic analysis techniques in a mixed-language clinical context.

Main Methods:

  • Designed a conceptual model through terminology exploration and lexical analysis.

Related Experiment Videos

  • Created Concept Node (CN) definitions based on syntactic relationship patterns.
  • Implemented an automatic Information Extraction (IE) system utilizing the developed CNs.
  • Evaluated the system's performance on typical Korean medical text containing mixed languages.

    • Main Results:

    • The developed system demonstrated the feasibility of extracting information from mixed Korean and English radiology reports.
    • Concept Nodes (CNs) proved effective as extraction rules for this specific clinical domain.
    • Syntactic and semantic analysis techniques contributed to successful information extraction.

    Conclusions:

    • The proposed Concept Node (CN) approach is effective for information extraction in multilingual SNUH radiology reports.
    • This method offers a viable solution for analyzing complex clinical text data.
    • The study highlights the potential of advanced NLP techniques for improving clinical data management and research.